Differential Development and Reproduction of the German
Cockroach (Dictyoptera: Blatlellidae) on
Three Laboratory Diets
RICHARD A. COOPER
AND
COBY SCHAL
Department of Entomology, Cook College, Rutgers University,
New Brunswick, New Jersey 08903
J.
Econ. Entomol. 85(3): 838-844
(1992)
ABSTRACT Development of nymphs and oocyte maturation in adults were examined
in Blattella germanica (L.) reared on three commercial diets. Nymphs fed rat food developed significantly faster than nymphs fed two commercial dog foods. Similarly, oocytes
matured more quickly in adult females that were raised on rat food than in females raised
on dog food. Nymphal development and oocyte maturation were slower in insects that
were fed whole dog food pellets than in insects fed ground dog food, suggesting that
grinding the diet diminishes a mechanical barrier in whole dog food pellets. Comparisons
of processed canine food and unprocessed canine food indicated that diets were significantly inferior after the steam extrusion process. Rat food is normally not subjected to such
conditions and both processed and unprocessed rat food are equally suitable for B. germanica. The significance and implications for comparative studies with cockroaches are
discussed.
KEY WORDS
Insecta, Blattella germanica,
NUTRITIONALFACTORS can exert profound
short- and long-term effects on the development
and reproduction of insects (see reviews by Bernays 1985, Dadd 1985, Reinecke 1985, Waldbauer & Friedman 1991). Although the general
nutritional requirements of most insects are similar, the optimal sources, types, and proportions
of nutrients vary widely among species and
among different developmental and reproductive stages (Dadd 1985).
Research on the nutritional requirements and
the regulation of feeding in cockroaches has generally included minimally defined diets that allow for the addition, deletion, substitution, or
alteration of the proportions of nutrients (e.g.,
Gordon 1959). However, most work on cockroaches does not examine nutritional factors directly; commonly, more crudely defined diets,
including various laboratory foods, are used for
maintenance of colonies. The sources and proportions of component nutrients vary widely in
these diets, and constituents other than the essential nutrients may act as growth factors or
affect the texture, palatability, and digestibility
of the diet, changing its utility to the insect
(Vanderzant 1974, Bernays 1985, Dadd 1985,
Slansky 1982, Slansky & Scriber 1985). Despite
the fact that different investigators use different
commercial or specially formulated diets, the
tacit assumption is made that use of different
diets does not preclude comparisons of results
from different laboratories. Dog food and rat food
are nutritionally complex, relatively adequate for
0022-0493/92/0838-0844$02.00/0
© 1992 Entomological
nutrition, reproduction
maintaining colonies of cockroaches, and routinely used in many research laboratories.
Herein we examine differences between these
diets to test this assumption.
Some insects can effectively withstand short
periods of starvation or nutrient limitation, but
most exhibit altered rates of development and
reproduction under such conditions (see Engelmann 1970). The German cockroach, Blattella
germanic a (L.), is highly sensitive to both food
deprivation and specific nutrient limitation, particularly proteins (Noland & Baumann 1951;
Haydak 1953; Kunkel 1966, 1981; Roth & Stay
1962; Hamilton & SchaI1988). We therefore hypotheSized that if various laboratory diets differ
nutritionally they might result in differential development and reproduction in this cockroach.
Furthermore, our preliminary observations indicated that nymphal development in B. germanica was slower on whole pellets of laboratory
food compared with ground food, suggesting a
mechanical barrier to feeding.
In this report we document that several life
history parameters of the German cockroach are
sensitive to changes in the diet and result in
delayed adult emergence and reproduction on
some diets. We also elucidate a mechanism for
the deficiency of some commercial diets. Finally,
we discuss the implications of these findings for
physiological and biochemical studies that use
these diets interchangeably with the German
cockroach and other species of cockroaches.
Society of America
June 1992
COOPER
&
SCHAL: DIETS
AND COCKROACH
DEVELOPMENT
839
100
Materials and Methods
Gl
U
C
Gl
C)
whole
small
large
]
Rat
chow
Insects. German cockroaches were obtained
medium
80
from a stock colony raised on whole Purina Rat
Gl
Chow (number 5012, Purina Mills, St. Louis,
E 60
Gl
Mo.) and water at 27 ± 0.5°C and a photoperiod
Dog
of 12:12 (L:D). Cockroaches were collected from tfl.
chow
Gl
40
the stock colony at various developmental stages
~
whole
and placed on one of several diets on which
as
nymphal development and oocyte maturation
"5 20
E
were examined. All experiments were conducted
:J
at 27 ± 1.0°C, 50% RH, and a photoperiod of 0
0
12:12 (L:D).
90
50
60
70
40
80
30
Diets. We suspected, based on preliminary reTime
(days)
sults, that the rate of nymphal development
would be affected by the size of the food items.
Fig. 1. Cumulative percentage emergence of adult
To test this hypothesis, dog chow (number 1780, females raised on either rat or dog food ground to
various extents. Neonate nymphs were fed either rat or
Ralston Purina, St. Louis, Mo.) and rat chow
dog food. Small, medium, and large food sizes refer to
were ground with a mortar and pestle and sieved
0.7-1.0, 1.7-2.6, and 2.6-4.0 mm respectively, as desequentially to obtain the following three ranges
scribed in the Materials and Methods. Five groups of
of particle sizes: 0.7-1.0 mm, 1.7-2.6 mm, and
20 nymphs each were used for each treatment.
2.6-4.0 mm.
Results of the nymphal development study
suggested that differential growth might be reTo avoid deleterious effects ofisolation (Gadot
lated to the differing degrees of compression and
et al. 1989), newly emerged adult females from
extrusion experienced by different diets during
various experimental treatments were placed in
production. Rat and dog foods in the form of meal
15 by 2 cm petri dishes in groups of two to five.
before processing are relatively dry, with a sim- On day 7, the basal oocytes of each female were
ilar fat content of 4.5 and 4% respectively (D. measured with an ocular micrometer in a dissectHopkins, personal communication). Rat chow
ing microscope.
meal is pressed into a cylindrical shape with two
Nymphal development and oocyte maturation
cut ends and a dry smooth surface. However, dog were also compared in cockroaches fed either
chow meal is steam extruded under high presunprocessed canine or rat chow or the processed
sure and temperature; during this process, an form of each diet including dog chow, which was
additional layer of 5% animal fat is sprayed onto ground to 0.7-1.0 mm.
the outer surface. The end product is a hard
Statistical Analyses. All results were analyzed
round pellet with a "baked" lipid surface and a either with analysis of variance and Gamestotal fat content that is twice that of rat food (D. Howell multiple comparison of means with unequal variances (P S 0.05 [Super ANOVA 1989])
Hopkins, personal communication).
The hard, greasy outer surface of dog food or with Student's t test (P S 0.05).
might act as a mechanical barrier, especially to
young nymphs. Because we were unable to obResults and Discussion
tain unprocessed Purina Dog Chow, we used
Nymphal Development. Almost twice as many
Purina Canine Chow (number 5006, Purina
female nymphs that were fed rat food reached
Mills), which is similar in nutritional composition and processing to Purina Dog Chow. To adulthood, compared with nymphs fed dog food
(Fig. 1). Nymphs that were raised on rat food also
control for the different fat contents of the processed diets, the fat contents of canine and rat eclosed significantly earlier than those fed dog
food (Fig. 1, Table 1). However, newly emerged
meals were raised to 9.0% in some experiments
by first dissolving animal fat (Purina Mills) in adult females attained similar mean body mass
on both diets (Table 1), suggesting that delays in
methanol and petroleum ether (1:3 by vol) and
eclosion in females fed dog food were from a
then impregnating it into the meal.
slower increase in body mass. In all dietary treatNymphal Development and Oocyte Maturaments, neonates that were fed rat food gained
tion. Groups of 20 newly hatched nymphs were
weighed, placed together in petri dishes (15 by 2 significantly more mass during the first 5 wk of
development than nymphs that were fed dog
cm), and provided with water and either whole
food (Fig. 2). These differences were apparent
or ground dog or rat food. On a weekly basis,
within 1 wk and continued to increase throughnymphs were counted and weighed, food and
out the remainder of nymphal development.
water were replaced, and debris was removed.
Grinding the food pellets significantly imThe time until eclosion and imaginal weights
proved the usefulness of dog food, but not rat
were also recorded.
•..
-
~~~L]
840
JOURNAL
OF ECONOMIC
Table I. Day of eelosion and adult weight (:i ± SEM) of B. germanica
Rat Chow ground to various mesh sizes
females raised on either Purina Dog Chow or
Dog chow
Mesh size, mm
0.7-1.0
1.7-2.6
2.6-4.0
Whole pellets
Vol. 85, no. 3
ENTOMOLOGY
Day of eelosion
n
26
29
26
14
47.3
45.9
46.5
66.4
±
±
±
±
Rat chow
Mean weight, mg
n
±
±
±
±
73
42
42
46
0.6a
0.7a
0.7a
2.7b
63.8
63.3
64.5
63.5
0.8a
0.9a
0.8a
1.2a
Day of eel os ion
39.1
39.4
38.7
38.2
±
±
±
±
0.3a
O.4a
0.8a
0.2a
Mean weight, mg
63.6
65.9
64.0
63.9
±
±
±
±
0.8a
1.1a
0.8a
0.9a
Within columns, means followed by the same letters are not significantly different (P > 0.05, ANOVA and Games-Howell
multiple comparison of means [Super ANOVA 1989]). All comparisons of the mean day of eelosion for respective mesh sizes of
dog food and rat food are significantly different (P < 0.05, Student's t test). All comparisons of the mean weight for respective mesh
sizes of dog food and rat food are not significantly different (P > 0.05, Student's t test).
food, for B. germanica female nymphs. The size
of the rat food diet had no significant effect on
mean body mass of5-wk-old nymphs (Fig. 2), the
percentage of females eclosing successfully, or
their mean day of eclosion (Fig. 1, Table 1). Differences in the mean day of eclosion were <1.2 d
(=3% of the mean), and had no consistent relation to the size of the food. In contrast, nymphs
that were provided whole dog food pellets
gained significantly less mass after 5 wk of development than nymphs that were fed ground
dog food (Fig. 2). Whole dog food pellets also
retarded nymphal development by =43% compared with the developmental period exhibited
by insects that were fed ground dog food (Fig. 1,
Table 1); whole dog food also reduced the percentage of eclosing females by =50%. However,
there were no significant differences in all three
parameters among nymphs fed dog food ground
to various mesh sizes (Table 1). Although developmental rates were improved by grinding the
dog food diet, significant differences in the mean
day of eclosion persisted
between
equally
ground dog food and rat food diets (Table 1).
This supports our conclusions that a mechanical
barrier in whole dog food pellets is responsible
for the slower growth and development
of
60
a
50
Ci
g 40
:cCI
b
whole
]
medium
large
small
Rat
chow
~~:~um]
small
Dog
chow
30
'Qj
:;: 20
C
whole
10
o
o
2
3
456
Time (weeks)
Mean wet weight (± SEM) of nymphs on
Fig. 2.
the 5th wk of development. Insects were treated as in
Fig. 1. Different letters indicate significant differences
among treatments (P < 0.05; ANOVA and GamesHowell multiple comparison of means [Super ANOVA
1989]).
nymphs, and that rat food supports nymphal development significantly better than does dog
food, regardless of the mesh size of the diet.
Reproduction.
At eclosion, females attained
similar mean body masses on all dietary treatments, including whole dog food pellets (Table
1). However, these females exhibited different
reproductive rates, as indicated by their basal
oocyte lengths. Basal oocytes were significantly
larger in 7-d-old adults that were reared on rat
food than in females reared on dog food (Fig.
3A). These differences persisted on all mesh
sizes of the two diets, but were greatest on whole
pellets. Within either diet, tlle size of food particles did not affect oocyte maturation except in
females raised on whole dog food. In the latter,
the basal oocytes were Significantly smaller than
in all other treatments (Fig. 3A).
Differences in oocyte maturation between females diminished when they were fed rat food
until the last instar or until adult emergence (Fig.
3B and C). Significant differences persisted only
when last instar or adult females were switched
from rat food to whole dog food. These results
suggest that reserves acquired during earlier instars may be mobilized
during
the first
gonotrophic cycle in B. gerrnanica females that
are switched to deficient diets.
Comparison of Processed and Unprocessed
Diets. Differences between whole and ground
dog food suggest that physical attributes such as
texture, hardness, and homogeneity might be involved. However, because the inferiority of dog
food compared with rat food can be ameliorated
only in part by grinding, the processed dog food
may also be nutritionally deficient.
To test this hypothesis, we obtained unprocessed meals of these diets. Because we were
unable to obtain unprocessed Purina Dog Chow,
we used Purina Canine Chow, which is similar
in nutritional composition and processing to Purina Dog Chow. During the commercial extrusion (dog chow and canine chow), the quality of
these diets for B. germanic a might be altered.
Nymphs that were fed extruded ground canine or
dog foods weighed less, reached the adult stage
later, and experienced
higher mortality ilian
June 1992
COOPER
Rat chow
&
0
SCHAL: DIETS AND COCKROACH
841
DEVELOPMENT
Dog chow
A. FIIlIllkls swllChed 10 dl&lS as newty halC/1ed nymphs
2.0
.-
••
a
a*
a·
1.5
1.0
0.5
E
8 .. Females
§. 2,0
.c
~
1.0
U
<3
swI1chIId 10 dl&lS as IasIlnslal
•
nymphs
8
35
1.5
..!!!
~
•
0.5
2.0
processed
unprocessed
unprocessed
+
+
ground
IaI
Fig. 4. Mean wet weight (± SEM) of nymphs in the
5th wk of development. Newly hatched nymphs were
fed rat or canine meal (unprocessed) with or without fat
added (see Materials and Methods); or rat, canine, or
dog food (processed) that was ground to 0,7-1.0 mm.
Different letters indicate significant differences among
treatments (P < 0.05; ANOVA and Games-Howell
multiple comparison of means [Super ANOVA 1989]),
1.5
non might explain the reduced growth and reproduction of cockroaches on this diet.
0.5
Steam extrusion may affect diets in at least two
0.7 -1.0
1.7 - 2.6
2.6 - 4.0 WHOLE FOOD
ways. First, by producing a "baked" lipid exteFood size
rior, this process may introduce a mechanical
Fig. 3. Oocyte size in 7-d-old adult females that barrier that reduces food intake and retards dewere fed rat food or dog food (whole pellets or ground velopment and reproduction.
This hypothesis
to smaller mesh sizes). Nymphs and adults were either
was supported by the accelerated development
maintained on rat food or switched to dog food as (A) of nymphs that were fed extruded ground dog
newly hatched nymphs, (B) as last instars, or (C) as food compared with extruded whole pellets.
newly emerged adults. Bars indicate SEM. Different
letters within the same food indicate significant differ- Similar effects of texture and hardness of diets
have been reported in other insects (Vanderzant
ences among food sizes (P < 0.05, ANOVAand GamesHowell multiple comparison of means [Super ANOVA 1969). Second, steam extrusion may alter the
1989]). *, Significant differences between diets of a composition or digestibility of the diet. Nymphs
that were fed unextruded canine diet exhibited
particular mesh size (P < 0.05, Student's t test).
significantly faster growth than nymphs fed extruded ground canine diet. Extrusion subjects
the diet to high temperatures
and pressures,
nymphs fed processed (pressed) rat food (Fig. 4 which may have adverse effects on dietary constituents including vitamins, starches, and amino
and 5, Table 2). Although B. germanica females
performed significantly better on extruded ca- acids.
Relationship to Other Studies. These results
nine food than on extruded dog food, nymphs
have important implications to numerous studies
that were fed unprocessed forms of these diets
on the life history, physiology, toxicology, behavattained similar weights (Fig. 4) and exhibited
ior, and biochemistry of cockroaches and possisimilar adult emergence patterns (Table 2). Conbly other insects. For example, in Spodoptera
versely, nymphs that were fed either processed
or unprocessed rat food exhibited identical de- frugiperda (J. E. Smith), changes in glutathione
S-transferase
activity as a result of dietary
velopmental and mortality patterns. The addichanges can affect tolerance of organophosphate
tion of animal fat to unprocessed diets, which
insecticides (Clark 1989). Whether different labresulted in diets of similar compositions to the
oratory diets have similar effects on the German
processed diets, did not alter nymphal developcockroach remains to be determined.
mental rates or the number of adult females that
Different results from different laboratories on
emerged (Fig. 4 and 5; Table 2). These results
indicate that the commercial extrusion process in the same cockroach species have generally been
attributed to different environmental conditions
production of canine diet diminishes its suitabil(particularly temperature), different experimenity for B. germinica. Although this could not be
tal protocols (such as anesthetics), and different
investigated with dog chow, a similar phenome1.0
842
JOURNAL
OF ECONOMIC
.......•...
--~-
-
--•
100
CI)
---0--
(,)
c:
CI)
C)
80
E
CI)
60
Vol. 85, no. 3
ENTOMOLOGY
•••
CI)
rat, unextruded meal
rat, unextruded meal + fat
rat, processed ground
canine, unextruded meal
canine, unextruded meal + fat
canine, extruded ground
dog, extruded ground
~
o
CI)
>
.•••
40
E
20
-as::s
::s
o
o
30
35
45
40
55
50
Time (days)
Fig. 5. Cumulative percentage emergence of adult females raised on either rat or canine meal (unprocessed);
or processed rat, canine, or dog food that was ground to 0.7-1.0 mm.
strains that may be differentially responsive to
the experimental
manipulation.
However, dietary differences are rarely considered. Despite
the fact that different diets are used in various
laboratories,
the diets are presumed
to be
equally adequate for growth and reproduction of
cockroaches. Based on the present work, we suggest that some discrepancies
in the literature
might be explained in part by the nutritional
state of the insects. Regulation of the cockroach
corpus allatum (CA) may be taken as an example.
In all cockroaches studied to date, the CA,
which synthesizes juvenile hormone (JH), is restrained to varying degrees by signals from the
brain (for reviews, see Feyereisen [1985] and
Tobe & Stay [19851). Specific cues such as those
from mating, feeding, and various social stimuli
can, together or independently,
lift the brain inhibition on the CA (e.g., Cadot et al. 1989).
Table 2.
unprocessed
Day of eclosion and emergence
laboratory diets
Form of diet
Processed (0.7-1.0 mm)
Unprocessed
Unprocessed + fat
Therefore, denervation of the CA has generally
been considered to mimic these species-specific
stimuli. In all cockroach species that have been
examined, CA-denervation potentiates JH synthesis and oocyte maturation beyond the normal
levels exhibited by females that experience all
the appropriate cues. Our data suggest that certain commercial diets may he deficient for some
cockroach species, possibly resulting in varying
degrees of CA inhibition in normally grown females, as shown by suppressed oocyte maturation.
This hypothesis pertains to a current dispute
in the literature. Weaver (1984) showed that CAdenervation
of mated American cockroaches,
Periplaneta americana (L.), reduced both JH
synthesis and oocyte growth. These results contradicted Pipa's (1982) conclusions
that CAdenervation
potentiates
oocyte maturation in
weights (% ± SEM) of B. germanica
Rat chow
n
Day of eelosion
25
26
26
37.6 ± OAAa
37.9 ± OAAa
37.7 ± OAAa
females raised on processed
and
Dog chow
Canine chow
n
Day of eelosion
n
Day of eelosion
23
27
26
42.0 ± 0.7Bb
37.5 ± OAAa
37.7 ± OAAa
19
45.9 ± 0.6C
NA
NA
Within columns. means followed by different letters are significantly different (P < 0.05, ANOVA and Games-Howell
comparison of means [Super ANOVA 1989]). Within rows, means followed by different capital letters are significantly
(P < 0.05, ANOVA and Games-Howell multiple comparison of means). NA, not available.
multiple
different
June 1992
COOPER
&
SCHAL: DIETS AND COCKROACH DEVELOPMENT
starved virgin females. In addition to the differences in experimental conditions between the
two research laboratories (outlined in Pipa
[1986]), our data with B. germanica suggest that
differences in diets may have contributed significantly to this discrepancy. In Pipa's laboratory,
cockroaches were fed dog food, whereas Weaver
provided them with a ground mixture of oatmeal,
dog food, peanuts, and yeast powder (17:10:4:1,
respectively). If the latter dietary mixture is more
adequate for P. americana, the CA of fed mated
females in Weaver's experiments would be more
disinhibited than those in Pipa's studies, explaining their conflicting results. Indeed, our
preliminary results indicate that JH release rates
are greater in the CA of B. germanica females
that were fed rat food than in females fed dog
food, and that the degree of disinhibition by
nerve transection is inversely related to the absolute (control) rate ofJH synthesis (C.S., unpublished data).
A second example is related to our observations that in the first gonotrophic cycle, the female German cockroach mobilizes nutrient reserves from earlier instars as well as those
acquired during the intensive preovulatory feeding period, and that superior nymphal diets may
buffer the effects of inferior adult diets (Fig. 3).
The effect of starvation on subsequent oocyte
development in P. americana was investigated
in four different laboratories (Kunkel 1966, Bell
1971, Weaver & Pratt 1981, Pipa 1982). In three
of the studies, females were fed ad libitum before oviposition of the first ootheca, after which
they were starved. In each study, however, the
diet was different as were the results. Kunkel
(1966) reported that starved females produced
four to five oothecae; in studies by Bell (1971)
and Weaver & Pratt (1981), starved females produced no more than three oothecae. The percentages of females ovulating a second, third,
and fourth time also varied by as much as 59%
among the three studies. Kunkel (1966) used
Ken-L Ration and Purina Rat Chow; Weaver &
Pratt (1981) used a mix of dog food, oatmeal,
peanuts, and yeast; and Bell (1971) used an unspecified laboratory diet. The possibility that nutrition may have affected these results was considered by Weaver & Pratt (1981), who suggested
that both the quantity and the quality of food
consumed before starvation may have affected
the amount of stored metabolites available to females. This notion appears to be supported by
Pipa's (1982) results demonstrating that <1% of
P. americana females fed Purina Dog Chow ovulate when starved upon eclosion.
Finally, several works report widely disparate
life history parameters in P. americana (see Roth
1981). Some of these differences can probably be
explained by the diversity of the diets, which
include potato cubes, raw meat, apples, and bananas, as well as various dog foods and lab diets
843
(Gould & Deay 1938, Rau 1940, Geir 1947, Willis
et al. 1958, Kunkel 1966).
Our results with B. germanica, together with
inferences from Periplaneta and other species,
sugg~st that the quality of laboratory diets may
directly or indirectly profoundly affect the interpretation of endocrinological, biochemical, and
toxicological studies. We strongly suggest that
detailed descriptions of dietary regimes be given
in such reports.
Acknowledgment
We thank D. Hopkins (Purina Mills) for providing
Purina rat and canine meals, Purina Canine Chow,
animal fat, and technical information on their commercial processing. Supported in part by grants from
USDA-CSRS (90-34103-5413) and the Charles and Johanna Busch Memorial Fund to C. Schal and the
Thomas J. Headlee Fellowship to R. A. Cooper. This
article is New Jersey Agricultural Experiment Station
publication D-08928-08-91, supported by State Funds
and by the U.S. Hatch Act.
References Cited
Bell, W. J. 1971. Starvation-induced
oocyte resorption and yolk protein salvage in Periplaneta americana. J. Insect Physiol. 17: 1099-1111.
Bemays, E. A. 1985. Regulation of feeding behaviour, pp. 1-32. In G. A. Kerkut & L. I. Gilbert [eds.],
Comprehensive
insect phYSiology, biochemistry
and pharmacology vol. 4, 1st ed. Pergamon, Oxford.
Clark, A. G. 1989. The comparative enzymology of
the glutathione S-transferase from non-vertebrate
organisms. Compo Biochem. Physiol. 92B: 419-466.
Dadd, R. H. 1985. Nutrition: Organisms, pp. 313390. In G. A. Kerkut & L. I. Gilbert [eds.], Comprehensive insect physiology, biochemistry and pharmacology vol. 4, 1st ed. Pergamon, Oxford.
Engelmann, F. 1970. The physiology of insect reproduction. Pergamon, Oxford.
Feyereisen,
R. 1985. Regulation of juvenile hormone titer synthesis, pp. 391-429. In G. A. Kerkut
& L. I. Gilbert [eds.], Comprehensive insect physiology, biochemistry and pharmacology vol. 7, 1st
ed. Pergamon, Oxford.
Gadot, M., E. Bums & C. Schal. 1989. Juvenile hormone biosynthesis
and oocyte development
in
adult female Blattella germanica: Effects of grouping and mating. Arch. Insect Biochem. Physiol. 11:
189-200.
Gier, H. T. 1947. Growth rate in the cockroach
Periplaneta americana (Linn.). Ann. Entomol. Soc.
Am. 40: 303-317.
Gordon, H. T. 1959. Minimal nutritional requirements of the German cockroach, Blattella germanica. Ann. N.Y. Acad. Sci. 77: 290-351.
Gould, G. E. & H. O. Deay. 1938. The biology of
the American cockroach. Ann. Entomol. Soc. Am.
31: 489-498.
Hamilton, R. L. & C. Schal. 1988. Effects of dietary
protein levels on reproduction and food consumption in the German cockroach (Dictyoptera: Blattellidae). Ann. Entomol. Soc. Am. 81: 969-976.
Haydak, M. H. 1953. Influence of the protein level
844
JOURNAL OF ECONOMIC ENTOMOLOGY
of the diet on the longevity of cockroaches. Ann.
Entomol. Soc. Am. 46: 547-560.
Kunkel, J. G. 1966. Development and the availability of food in the German cockroach, Blattella germanica (L.). J. Insect Physiol. 12: 227-235.
1981. A minimal model of metamorphosis: Fat body
competence to respond to juvenile hormone. pp.
107-129. In B. Ghovindan, S. Friedman & J. G.
Rodriquez [eds.], Current topics in insect endocrinology and nutrition. Plenum, New York.
Noland, J. L. & C. A. Baumann.
1951. Protein requirements of the cockroach Blattella germanica
(L.). Ann. Entomol. Soc. Am. 44: 184-188.
Pipa, R. L. 1982. Neural influence on corpus allatum activity and egg maturation in starved virgin
Periplaneta americana. Physio!. Entomol. 7: 449455.
1986. Disinhibition of oocyte growth in adult virgin
Periplaneta americana by corpus allatum denervation: Age dependency and relatedness to mating.
Arch. Insect Biochem Physiol. 3: 471-483.
Rau, P. 1940. The life history of the American cockroach Periplaneta americana Linn. (Orthop.: Blattidae). Entomol. News 51: 121-124, 151-155, 186189,223-227,273-278.
Reinecke, J. P. 1985. Nutrition: Artificial diets. pp.
391-419. In G. A. Kerkut & L. I. Gilbert [eds.],
Comprehensive
insect physiology, biochemistry
and pharmacology vo!. 4, Ist ed. Pergamon, Oxford.
Roth, L. M. 1981. Introduction. pp. 1-14. In J. Bell
& K. G. Adiyodi [eds.], The American cockroach.
Chapman & Hall, New York.
Roth, L. M. & B. Stay. 1962. Oocyte development in
Blattella germanica and Blattella vaga (Blattaria).
Ann. Entomol. Soc. Am. 55: 633-642.
Slansky, F. Jr. 1982. Toward a nutritional ecology of
Vol. 85, no. 3
insects. Proc. 5th Int. Symp. Insect-Plant Relationships, Wageningen.
Slansky, F., Jr., & J. M. Scriber.
1985. Food consumption and utilization. pp. 87-163. In G. A.
Kerkut & L. I. Gilbert [eds.], Comprehensive insect
physiology, biochemistry and pharmacology vo!. 4,
1st ed. Pergamon, Oxford.
Super
AN OVA. 1989. Abacus
Concepts,
Inc.,
Berkeley, Calif.
Tobe, S. S. & B. Stay. 1985. Structure and regulation of the corpus allatum. Adv. Insect Physio!. 18:
305-432.
Vanderzant, E. S. 1969. PhySical aspects of artificial
diets. Ent. Exp. App!. 12: 642-650.
1974. Development,
significance, and application
of artificial diets for insects, Annu. Rev. Entomo!.
19: 139-160.
Waldbauer,
G. P. & S. Friedman.
1991. Selfselection of optimal diets by insects. Annu. Rev.
Entomo!. 36: 43-64.
Weaver, R. J. 1984. Effects of food and water availability, and of NCA-l section, upon juvenile hormone biosynthesis
and oocyte development
in
adult female Periplaneta americana. J. Insect Physio!. 30: 831-838.
Weaver, R. J. & G. E. Pratt. 1981. Effects of starvation and feeding upon corpus allatum activity and
oocyte growth in adult female Periplaneta americana. J. Insect Physio!. 27: 75-83.
Willis, E. R., R. R. Riser & L. M. Roth. 1958. Observations on reproduction and development in cockroaches. Ann. Entomol. Soc. Am. 51: 53-69.
Received for publication
December 1991.
27 March 1991; accepted 5